Docs
Documentation
Everything you need to install, configure, and run kRouter in production.
Reference
API Reference
OpenAI-compatible /v1/chat/completions, /v1/messages, /v1/models endpoints exposed by kRouter on localhost:20128.
Architecture
How kRouter routes a single chat completion through 7 internal stages, from your IDE to the upstream provider and back.
CLI reference
Every flag the `krouter` command supports. Tray mode, port override, MITM toggle, log streaming, version pinning.
Concepts
Core concepts
Providers, combos, tiers, RTK, MITM mode, quota tracking — the five ideas that make kRouter work.
Providers
Connect to 62+ AI providers across OAuth subscriptions, free tiers, and pay-per-token APIs. Each provider takes seconds to set up.
Combos & fallback
Stack subscription → cheap → free into a single endpoint. When one provider hits a quota wall, the next takes over. No interruption.
Operating
Deployment
Run kRouter on a VPS, in Docker, behind PM2, or as a cloud worker. Production-ready configurations and environment variables.
Security
How kRouter handles credentials, secrets, certificates, and exposes itself to the network. Threat model included.
Troubleshooting
Real fixes for the problems people actually hit. OAuth refresh failures, rate-limit cascades, MITM certificate issues, dashboard freezes — all here.